OF SILOS and BUNKERS PDF

O F SILOS AKD BUNKERS
25
Discussion
Mr Leopold Brook introduced thePaper,inthe

absence of the
Author, with the aid of a series of lantern slides.
TheChairman said that the Author was rightinemphasizing
the
importance of foundat,ions, because silos were among those rare structures
whichwere frequentlyloaded to their full design load ; in most cases
relative settlement w-as what mattered and not absolute settlement.Fortunately, silos were a type of structure which could itself often be utilized
to redistribute foundation loads and so reduce relative settlement. Care
was not always taken with the foundation. A silo a t Algiers had failed
completely by sliding on a circular arc-there was a picture which showed
it leaning over a t about 45 degrees.
Could the Author give more guidance on wall thickness ? In the Paper
5 inches was mentioned as a reasonable thickness. How did
the duthor
select a thickness and reinforcement to make a wall weather-tight ?
With regard to formula (3), was the horizontal pressure pl always the
active or minimumpressure ? In deeper silos it might be considerably
more.
In the Paper, timber was dismissed because of its low tensile strength.
The Chairman suggested that it was not a question of low tensile resistance ; the tensile strength of timber was quite good, but old-fashioned
methods of jointing timber in tension were extremely weak. He thought
that modern laminated technique might offersome interesting solutions
in the const,ruction of silos. Reference was made in the Paper to abrasion,
and abrasion might put timber out of court, but it should be possible to
design satisfactorily a silo in timber, so far as strength was concerned.
He agreed with the Author that reinforced-concrete structures had an
unduly bad name for difficulty of alteration. It was often quite as diffiwas to alter a reinforced-concrete
cult t o alter a steel structureasit
structure.
Mr F. S. Snow, referring to the statementmade in the Paper regarding
the short life of timber and particularly to its lack of fireproof qualities,
said that there weresome 50-foot-high silos built of timber a t Trafford
Park, which had been bombed and had caught fire, but they had never
actually burnt out ; once the timber had charred for Q inch on the outside, the remainder of the timber remained intact. There was a tendency,
especially where t,hick timber was used in large sizes, t o over-emphasize
the fact that itwas affected bp fire.
Had the Author utilized steel formwork in the sliding formwork for
theshutters completely,instead of timber,and, if so, what were the
advantages or disadvantagesof steel formwork compared withany other ?
Assuming that the sliding formwork had been constructed about the
silos, then when the form was moved upwards therewould be considerable
distortion inthe formwork causedby the heat and the water inconcret,e
the
Downloaded by [] on [06/07/16]. Copyright ICE Publishing, all rights reserved.
26
DISCUSSION
ON
THE DESIGN AND CONSTRUCTION
those forms might distort as much as

3 or 4 inches in any direction. Moreto twist,
over, in jacking up it would befound that theforms had a tendency
so that, starting with a 5-inch wall, the thickness might become 3%inches
in one place and 5Q inches in another. To overcome such difficulties, a
triangulated set of tie-rods could be put across the formwork with unions
in between, so that as the formwork rose it wouldbepossible to make
adjustments on those unions andpull the silos back to the designed shape.
Mr Snow commented on the value of loud-speakers, which had been
very useful for giving orders to the men working on silos. I n bad weather
or when working a t night, it was often not possible t o get a level on the
silo when jacking up, and it might then be as much as 1 foot out-he
thought Mr Broughton wouldconfirm that. It wouldbe necessary to
correct that during the day, and with the aid
of loud-speakers the situation
could be explained to the men.
When building a silo 80-90 feet high, the walls of which were of the
order of 5 to 6 inches thick, the roof had to be constructed, and unlessthe
scheme had been preparedbeforehand it would have been extremely
difficult and dangerous for the men to remove the formwork for the bins
and toconstruct shutteringfor the roof without going to thebottom of the
bins. Experience had shown that, by putting in two steel joists with an
extensible end (so that it was possible to leave those two joists in position
and extend them each side to overlap the wall) the existingformwork
could be used to do the concreting for the roof, leaving a hole in the centre
of each bin to get at the formwork to remove it. On completion of the
concreting, an instrument constructed in the
form of an umbrella was
dropped down throughthe holes of the bins and thenopened out inside the
silo, allowing the formwork to be supported, dismantled, and then passed
up through the previously mentioned hole in the centre of the bins. The
alternative method-and apoor one-was to leavea large hole in the
bottom of the hopper and drop the whole of the shuttering, leaving the
hopper bottom to be constructed a t a later period.
Mr Snow had first tried out theidea of a pictorial representation in the
time schedule (Pig. 2 2 ) on a block of the Guinness Brewery silos. That
added interest to the time schedule, particularly for a client who liked his
facts presented pictorially.
Pig. 23 showed the Victoria Docksilos,where a worlds record had
been created for speed of erection ; the date of that picture was 19 Sepwalls
tember, 1933. Fig. 24 showed the same silos 5dayslater.The
were 90 feet in height andwere constructed in 6 days 18 hours.
Mr H. H. Broughton said that hewould not confirm Mr Snows
remark that it was quite common t o have decks 12 inches out of level.
H e had never before heard of decks being 12 inches out of level.
Nowhere in the Paper did the Authorsay why hepreferred to use
hexagonal or other queer-shaped bins in preference to
the circular form.
Under the action of internal plessure, any bin, no matter what its shape,
Fig. 22
SEPTEMBEIL 1934
10th
17th
OCTOBER )P34
NOVEMaER 191+
DECEMBER I934
24th
W
P
Stel fixing
Concreting
Scheduler
Formwork
Sreel flxing
Coneretlng
Preparationof forms
and working drawings
Erearon of formwork
Steel fixing
Concreting
offormwork
Steel fming
Total contract
'W
PICTORILLL
REFRESENTATION
OF -E
SCEEDULB
JANUARY 1935
28
DISCUSSION
ON
THE DESIGS AND CONSTRUCTION
tended t o become circular,which was therefore the logical form. If

Fig. 7 were examined, it would be seen that there were bins of two different
sizes. The main circular bin had a volume of about 500 tons, whilst the
interspacebinhad a volume of about 100 tons.Byconstructing
walls
tangential and on the outside, an end bin or quarter bin wit'h a capacity
of about 50 tons could be obtained. If, therefore, a small parcel of grain
or mineral had to be stored, it could be put either in the end bin or in the
interspace ; that was the great advantage of circular bins over bins of
other forms. If the bins were all of one size, dead storage '' was created,
which present.ed great difficulty in the economical handling
of material.
Reference was madein thePaper,and
also inrecently-published
articles, to walls of fantastic thicknesses-4-44 inches. Mr Brought'on
regarded 6 inches as the minimumthicknessunder
any conditions for
moving formwork ; 7 inches should be used where possible, and perhaps
8 inches for larger bins.
Reference was also made in the Paper to the spalling of the concrete
due to the internal corrosion of reinforcing st8eel. That was not reinforced
concre.te, nor was it good engineering.
Thejack-rodspacinggiven
in thePaper
was 4-6 feet. Hehad
frequentlyused 12-13 feet. It dependedentirely on the arrangement,,
but it would not be advisable to regard 6 feet, as the maximum spacing of
the jack rods. The weight of jack-rod steel depended on the wall thickness
and rod spacing, and varied between 10-18 lb. per cubic yard of concrete ;
15 lb. was a good average.
He was sure that the Author wouldbe one of the first to welcome
criticism of thearrangementshownin
Figs 16. Figs 16 showed the
arrangement of the yoke and jack, and thehanging scaffold, together with
the method of supportingthe walings from the yoke. That diagram
represented the arrangement which Mr Broughton had used in about 1924,
and which had been abandoned in about 1929. It worked, but not so well
as the modern arrangement. Instead of the timber yokes, which weighed
297 lb. each, indestructible pressed-steel framesweighing 137 lb. were now
used. It had been found that the jaw clutch worked quite well so long
as the jack rod remained 1 inch in diameter, but no commercially-rolled
rodcould be relied on for constancy of diameter; if the diametergot
small the clutch went the wrong way and locked on the rod, so that the
clutchhad to be dismantled.Themodernclutch,
therefore, took the
form of a sleeve overthe jackrod and was connected to thejack rod, and in
the top there was a self-aligning ball and thrust bearing.Theexisting
jack screw was used, but sleeves were now used instead of jaw clutches.
That arrangement hadreduced the cost of jacking.
The hanging scaffold should not be supported by or hung from
the
yoke, because it was necessary to ret,ain the scaffold in position against
the side of the building when the yokes had been removed, and that was
was
not possible if the scaffoldwere attached t o the poke. The hanger
b.
29
OF SILOS AXD BUNKERS
supported, therefore, directly from t,he walings. With regard to the support of the walings, four angles were shown in Pigs 16. The objection to
that was that itinvolved a rigid arrangement. The walings were S cured
by bolts orscrews to theangles, and i t was almost a physical imposslbility
t o make any adjustment t o the forms a,s work proceeded.
Nr Snow had referred to walls varying in thickness. Walls did not
vary in thickness when the man concerned knew his job, because he corrected t,he forms as work proceeded. That was done by hanging walings
of the yoke, so that thetension went direct to the nut.
straight to the top
Mr Broughton was not certain, from reading the Paper, whether the
or not. Ruskin had
Author believed in havingarchitecturalfeatures
remarked that he would rather have been born blind than live to see some
of them.
Fig. 25 (facing p. 27) showed the type of yoke which was now used.
It had pressed steel legs. The clutch was shown clearly, andthe jack
screw. The wedgesgoing completely round the form wouldbe noted.
Concreting had not pet commenced ; something had happened, and the
draw of t,he staves had had t o be corrected by means of those wedges,
a
Wit,hout pilasters, the building shown in Fig. 26 wouldlooklike
packing case. That building had t,o be in close contact with an existing
building, and had been constructed using the wall of the existing building
as the outside shuttering until reaching the level AB, where the higher
part began. At that point a cut-off hadtaken place. Half theforms
had been left behind, a new side form had been introduced, and concreting had then continued. That was a silo of 17,000 tons capacity ; the bins
were 109 feet deep and there was 30.4 cubic yards of concrete per foot of
heighb-a total of 3,300 cubic yards. The tall building on the right was
172 feet high. The contact surface of the formsfor the concrete was
10,400 square feet ; the wall perimeter, therefore, was 2,600 feet. So far
as labour costs were concerned, it. was a pre-war job. For the forms,
fabricating, erecting, maintaining.anddismantling,the
whole of the
labour costs had been 23.4d. per square foot of form. If 23.4 uses could be
got out of the timber,it,would be seen that the form cost for labour had
been only Id. per square foot. The figureswhich he gave werecertified
costs on which payments had been made.
The jacking costs for the lower part of the building had been 4.4d. per
jack-rodfoot. On reaching the higher part t,he work had become more
difficult ; it wa,s a four-storey building, and the jacking cost had been
6-85d. per foot. The steel hoisting and placing had cost 39s. 10d.-about
S2 per ton. The, concrete had come on the site ready mixed. The hoisting
and placing had cost 2s. 6Bd. per cubic yard. The average progress had
been a t the rate of 11 feet per day, and the bins had been constructed in
10 days, whilst everything shown in the picture had been completed in
about 3 weeks.
The subject of moving formwork was of sufficient importance to justify
e.
30
DISCUSSION
ON
THE DESIGN
AND
CONSTRUCTION
more consideration than had been given to it in the past. The equipment
might be rather costly, but to those who were interested i n construction,
he suggested that the contractors should establish a pool among themselves. Theremight be a poolof,
say,six-hundredjacksandother
fittings necessary, and those who got contracts would draw on thg. pool,
instead of having so much dead material in store year after year.
Mr G. P. Mannings first commentrelated to Airys formula. If
careful regard were givento the premises on which Airy based his results,
it would be seen that therewas a serious discrepancy, and it was surprising
to find that anyone still used Airysresults.
With regard t o sliding shutter work, the method described in the Paper
was not the only one. For example, in one contract the contractor had
a lot of old steel scaffold tubes and a large number of chain blocks, and
had built the scaffold tubes into the walls as he went up, hanging the
chain blocks from them and hoisting the shutters up continuously by that
means. The Author had said that theformwork was constructed for both
sides of the walls. That was not necessarily the case with moving shutter
work, and the inside shutters could be slid and the outside ones brought
up hand over hand. It was possible to slide a single-sided shutter.
The Author had suggested 4-6 feet for the spacing of the jack rods.
In Mr Mannings view, a normal spacing on a straight-forward job would
be about 7 feet, in which case each jack rod supported 56 square feet of
shutters, in addition to a certain amountof decking. It was stated in the
Paper that the shape of the silo should be designed to suit the formwork.
That was so, but it was also necessary to arrange the steel, if sliding shuttering were to be used, so that it could be easily passed underneath the
yokes. That was onedrawback to the hexagonalshape, where the bar
came up the side and turned a t a n angle. So far as the rate of sliding
was concerned, the figures given in the Paper appeared to be on the low
side. A fairly low average for a big job was 9 feet in 24 hours.The
fast,est slide of which he had any personal knowledge was 12g inches in
1 hour and 16 feet in 24 hours. Under really good conditions i t should
be possible to maintain a rate of 1 foot per hour. The
largest figure of
mhich he knew was 72 cubic yards of concrete per footof height, but there
should be no difficulty in sliding jobs two 01three times that capacity.
The highest slide of which he knew was 140 feet, but there should be no
difficulty in reaching 200 feet,. He was not thinking of chimney work but
of silo work as such. The thickest
wallwas 18 inches, and the thinnest
silowall 5 inches. TheAmericansdid notfavour sliding anything less
than 7 inches ; in Britain one should certainly not slide anything less than
6 inches. Mr Snow had already raised the question of how thick a 6-inch
wall was. Those who had seen sliding work could perhaps give a guess.
Mr Snow had also asked whether anybodyhad triedusing steel shutters
on sliding work. Mr Manning said that it had been done, but the application had not been of sufficient interest to give details of it.
OF SILOS AND BUNKERS
31
On the question of the design of silo bottoms, all silo bottoms were
really the same type of design, as shown in Pig. 27, the shaded space being
filled withmassconcrete.Theonlyvariation
was in the height, h. In
a very small silo it p i d to drop the slab so that the small hopper disappeared, but in a very large one it might pay to lift itso that the filling
disappearedand it was practically all hopper. For the normal bin of
12-14 square feet it usually paid to put the slab abouthalf way down.
It was stated in the Paper that the circular shape was more costly to
construct than the hexagonal. That statement was only partly true, over
a restricted range of sizes. In Britain the size of each individual bin in a
battery of silos was fixed by the miller and by milling considerations. The
miller wanted to handle his wheat in batches of 100, 200, or 300 tons, and
Fig. 21
so the engineer made the bins of the size required. In countries such as
Canada,however,whereenormousquantities
of grain were merelystored
until a ship arrived to take the grain away, the circular necked-out shape
was undoubtedly the cheapest.
The American and Canadian silo installations were all very large circular bins.
Mr I. Hey said that he had found a reluctance, particularly in Britain,
t o apply the sliding-form methodt o jobs equivalent to the Authors 3,000ton bunker. The
casewas clearly made out for sliding forms for large
grain-storage silos as used in North and South America. The Author had
stated that the hexagon was the best shape if a number of compartments
for diameters
were required to form a honeycomb inplan, but admitted that
larger than about 18 feet the circular bin was preferable, where the interspace silos could be used. Why should not the inter-space be used in all
cases ? Strangely enough, it was in Britain that engineers questioned the
desirability or advisability of using the inter-space in a run of circular
silos. Jn grain-storage silos where the loads ran up to 1,000 tons in big
bins, there was, as Mr Broughton said, 20 per cent. of the load in the interspace bins, and half that in theend bins. That was the universal practice ;
32
DISCUSSION ON THE DESIGN AXD CONSTRUCTION
it was only in Brit'ain-or with materials other than grain-that the interspace was not being used, and then it was said that square or hexagonal
bins were more economic than round bins. It was a matter of simple
arithmetic.
He was pleased that Mr Manning had called attention to the type of
hopperconstruction shown in Pig. 27, even though i t meant building
columns from the main slab and st,arting thesliding shuttering higher up,
rather than building the columns and going right through. It was not
possible to go ahead and complete the plan, because the contractor came
back t o build his hoppers and was a long time on those hoppers. With
the slabformation.that difficulty was avoided. Mr Hey thought that
7 metres was an economical size for circular bins of up to 30 metres in
height. In the design, the fact,ors relating to the progress of the job had
to be takeninto consideration. The basis of theadvantage, of sliding
formwork was that by a fair amount of preparation and organization time
was saved on construction. The steel mouth-pieces were easily fitted, but
should not be fitt'ed t o ternplated bolts.
I n Figs 2 a type of construction of hopper bottoms for sticky materials
was shown. It was known from experience that the best answer to the
problem o i handling st,icky materials was t o have t'wo vertical sides. With
two vertical sides he would not say t,hat arching would not occur, but
t,here would be no archingoff a vertical side ; vertical sides would guarantee
the best natural flow. That gave rise to a problem in the slope, but it
was the most st'raight-forward answer and was far better than hanging
the hopper on big bins from the direction of the top of the column. He
had seen building formwork done in record time, 12 feet a day, and then
it had been necessary to wait many months before the floor was free to
start placing the machinery.
It was stated in the Paper that the design had to be correlated with
the machinery and equipment'. Mr Hey looked on storage as compensabion between two systems of transport, from rail to ship or ship back t o
mil, or as compensation between sect'ions of a process, but always in the
sense of compensation. The Canadian and American grain siloswere
compensation as between producer and user and those forms of transport.
The second factor was that, even when use was made of the best methods
of constructing silos or bunkers, they were st.ill very expensive ; therefore
the mechanical handling or machinery process should be the first consideration, because storage was really a secondary function ; i t was the
machinery and plant and process which deteImined the earning capacity.
Simplicity of design for the use of sliding forms would give a good line
and contrast and a good-looking job.
So far as the thickness of the walls was concerned, he advocated a
thickness of 6 inches to give protection against t,he weather, but they had
to Fe good walls. He did not like t o go down lower than 5 inches, because
i t was not possible to ensure a proper cover on the steel ; but it coruld be
i
I
Fig. 30
//
33
time, and there were numerous jobs where it had been done and which
were very satisfactory. How Mr Snow came to have so great a variation
in the thickness of the wall he did not know ; that had not been his own
experience. Perhapsthe formwork had beenweak somewhere inthat
design.
Sliding fornlwork applied to the construction of big grain silos was
well illustrated by the example shown in Fig. 28, which was in the Argentine, and had a storage capacity of 140,000 t,ons, was 92 feet high, with
bins of 18 feet in diameter. It had been done in three lifts, and each of
those two blocks had taken S days to build. The receiving house on the
right was part of the big building. There were bins in between the main
floors, which had been stopped for several hours because they introduced
bin bottoms for dividing t.he bins, bottoms for elevator legs which did not
runrightthrough,andthere
were small walls for the elevators. The
at,oppages had been quite frequent,, but the 60 feet of those bins had been
carried out,fromMonday
afternoon toFriday morning, working with
squads of men of t,wenty-two different nationalities. Since 1932, when it
had been h i l t , i t had handled 1,000,ooOtons of grain a year,andin
October 1947 it received and shipped 200,000 tons of grain in thecalendar
month, which he thought was a record.
Fig. 29 showed a 150,000-ton silo in Buenos Aires ; it was possible to
feedfive ships simultaneously from that plant. Pig. 30 was an example
of moving formwork for a 60,000-ton silo, and Fig. 31 was another view
of the same elevator.
Mr F. G. Etches, dealing with the aesthetics of design, observed that
i t was inferred in the Paper that engineers did not consider sufficiently
the appearance of their structures. Whereas that might have been true
20 years ago, he did not think that it
applied to-day. It was certainly
t,rue that both bunkers and silos needed special thought and care in planning, because large surfaces, unbroken by window or door openings, and
usually without an interesting surface texture, presented certain problems.
That was especially t.rue of silos which were situated near a dock or on
the banks of a waterway, which t,ended to exaggerate their height and to
makc then1 appear out of proport,ion to their surroundings.
One solution was to attempt to create a simple self-contained and
balanced structure which did notdominate its neighbours. Mr Etches
described a nledium-sized silo in which small projecting piers, which were
nladc visihle by the shadows they cast, wereused to give form to the
building. In t.ha,t building there wasno wasted space, and 90 per cent.
of the area, and therefore the volume, was available for useful storage.
Another silo, of the same general arrangement and overall size, had eight
octagonal bins. The shadows on the elevation gave a broken and rather
rest,l(:ss surfacc, whichhe thought was also shown by Figs 17-19 in the
lapt:r----thelatter were hexagord bins, but the sitme point was illustrated.
Rather less than 80 per cent. was available for storage, and that included
3
34
DISCUSSION ON THE DESIGN AND CONSTRUCTION
the small square bins which, for grain storage, were often an advantage.
They stored about one-quarterof the quantity held by the large bins, and
qualities of binned
gavemore flexibilit,y for differentquantitiesand
material,
A circular bin gave probably the most interesting surface, as revealed
byshadows. One such lay-out had 70 per cent. of useful storagearea,
althoughagainsmallbins
were used.Another was a variation of the
square bin, but was improved by the addition of semi-circular piers. If
some justification had to be made for those piers, apart from appearance,
they could enclose rainwaterdown-pipes.
Almost the whole area was
available for storage ; in addition, there was the advantage of small bins,
and the plan cut down the large span of morJtof the walls, cxcept the
outside walls, as shown in Fig. 32.
With regard to the quantity of steel necessary in the bin walls, which
was influenced by the bin shape, the cconomics of any particular lay-out
of bins had to be investigated for each case, but the quantity of steel
mcessary lent itself to somegeneral conclusions. Fig. 33 showed an
dealised curve for a bin of given size, with the quantity of steel plotted
gainst the number of bin sides. The lower part of the curve showed the
tee1 required for ring tension, and was constant for all regular shapes of
in. Mr Broughton had made the point that all bins tended to become
rcular. The upper part of the curve from the dotted line upwards, to
e same scale, showed the steel required to resist bending stresses. That
tuld be maximum for a squarebin,and
was reducedinversely as
.
: square of the bin size ; it, was equal to zero for a circular bin.
The
11 of the t w o ordinates (the total vertical ordinate) would be a measure
;he total quantity of steel needed for any given set of conditions. The
:l required in the hexagonal bin was considerably less than thatfor the
are bin of the same overall size ; in other words, the curve was steep
35
to the left ; but the saving 'was likely to bo very much less for a greater
number of sides.
Fig. 33
COMPARISONOF HOBIZONTAL
WALLSTEELBEQUIBED IOR A GIVEN SIZEOF
BIN FOR VARIOUS GEOMETRIOAL FORME
Mr R. H.Squire questioned the validity of formula (3) given in the

Paper. If the slope of the bottom were as shown, and a small unit area
were considered (see Pig. 34), thevertical force (P,) acting on that
area wouldbe wh cos 8. Thehorizontal force (Pa) wouldbe equal to
1 - sin+
t case, and p , for
,wh.____ sin 8, or substituting p for wh in the h
1 sin
1 - sin+
wh. ___ in the second case, P, = p cos 8 and Pa = pl sin 8. Those
+ +'
1+&4
were forces andnot pressures, and so the dotted line represented the
resultant (P), PN being the normal and PT the tangential component.
In the Paper there
was no mention of PT, but it was not possible to
ignore it.
By inspection of Fig. 34,
PT=PVSh8-PpaCOS8
PN=PVCOs8+PH8h8
substituting p1 sin 8 and p cos 8 in those equations forP, and
P T = (p-pl) sin 8 COS 8 ; and
PN= p cos2 8 f p sin2 8.
4
P,,
36
DISCUSSION ON THE
DESIGN
AND CONSTRUCTION
The latter was the expression given in the Paper, and the former was the
expression for the tangential component.
He emphasized the fact that the tangential force had to be t,aken up
somewhere, and the only point where it could be taken up was by friction
being the
on the bottom, so that the angle M could not exceed #-that
friction angle betweenthe material and the bottom. Therefore formula(3)
PT
was only valid for obtaining the normal pressure when PN
+ tan$'.
If
that were greater than tan$' it would be necessary to make an adjustment

somewhere. Clearly it was not possible to adjust P,,the dead weight of
the material,which might or might not be reduced by friction on the upper
Fig. 34
Fig. 35
parts of the walls, according to whether or not a deep bin or deep silo
formula was being used, but if, having taken Rankine's or some other value
for the horizontal pressure and the friction on the bottom into consideration, the material still wanted to slide down, it wouldbe necessary t o
increase the lateral pressure ; that could make a considerable difference.
A solution couldbe obtained graphically by makingthe angle tc equal to $'
and getting a revisedvalue for P,, or by equating
!?? to tan $', and

p,
obtaining mathematically the value of p l .

With regard to the effect of unequal, unsymmetrical bins or unsymmetrical loading, using the Author's formula, the pressure on the bottom
would bedependent on the depth h (see Pig. 3 5 ) , so that at point A
there would be the same normal pressure on each side ; the pressure on
the vertical plane through A taken one way, however, would be different
from the pressure the other way, one being a positive surcharge and the
other a negative one, and also the sliding effect was almost always present.
37
If the friction angle was going to hold the material against sliding, there
wouldbe a very sluggish discharge, if there were any discharge a t all.
Reference was made in the Paper to the use of glass or similar lining for
the bottom to improve the discharge. In that case, he thought that the
normal pressure which would be taken on the bottom of the bin would not
be the same as if it were a concrete surface.
*** Mr Adolf Fruchtlander thought that the remarks in the Paper
regarding the pressure calculations and two standard formulae of Janssen
and Airy gave the impression that those formulae were more of an
empirical than a theoreticalnature ; that was not so. The basicconstants hadbeen determined, as in all theories of structures, byexperiment.
Jansscns formula had beendeveloped about 50 years agoon a purely
theoretical basis ; it had later been checked on models and buildings by
engineers, such as Luft, and found to be in agreement with the results of

the experiment)s. Since then the formula has been accepted as standard
for that branch of engineering.
Mr Fruchtlander then described briefly the theory of the fornlula in
question. Using the same notation as that given inthe Paper, hut
replacing thc friction coefficient tan C$ by p, the formula bccanlc :
In Pigs
36,
A.dp denohed t,lle increment of vertical pressure,

usA.dh
,,
,, weight of material,
pp1 U.dh
>,
,, side-wall pressure due to friction
between material and wall.
The equilibrium equation for those forces acting on an element a t a
depth h was :
A.@ = wA.dh - pp1U.dh. . . . . . (1)
t,
3,
***
This contribution was submitted in writing.SEo. I.C.E.
38
DISCUSSlON ON THE DESIGN ANI) CONSTRUUTION
Therefore
9dh -
IO
- pp1-U
A " " " "
(2)
P1
Putting - = k (constant), or dpl=k.dp, equation (2) becalm :
a24
U
- = kw - p*.k-p1
. . . . . .
A
&l
(3)
The solutivu of that equation was :

e--Pkh"/A
. . . . . .
(1)
which could easily be verified by differentiating equation (4)with respect;

to h. Equation (4)could also be written :
. . . . . . . (5)
where
= e-#hulA,
and
U pkh
log N = - __
A ' 2.303'
Fig. 37
'B
V
III practice, for a depth h more than 10-15 feet, N was a large number
approached unity.
Equations (4)and ( 5 ) therefore reduced to :
Aw
Maximum p - -
. . . . . .
If the curve shown in Fig. 37 represented the pressure values according

t o equations (4)and ( 5 ) ,the vertical line the asymptote to the curve for
h=co (according to equation ( 6 ) ) ,and the sloping straight line the presawes according to Rankine's formula p,=wh tan2 (45-4$), it would, in
39
general, be practicable to use the two straight lines for the design of the
walls ; that greatly simplified calculations, gave a good margin of safety,
and scarcely affected the economy of the design.
The Author, in reply, said that he very much regretted his inability,
due to illness, to introduce the Paper personally, and expressed his sincere
appreciation to Mr Leopold Brook for undertaking the task and
for showing
the lantern slides.
The Suthor was glad that theChairman hadcalled attention to failure
a
of the foundationsof a silo constructed inAlgiers. That typeof failure was
usually dueto thelack of consideration of the pressures on the stratabelow
the surface. Inthe case of isolated foundations a t a fairly generous
spacing, that was not of great importance, but in thecase of silo or bunker
foundationswhereahigh-intensitypressureoccurredoveracomparatively large area, the intensity of pressure on any weak strata below the
surface could be ten or twenty times higher than it would be under any
normal fourldation condition.
The thickness of a silo wall was arrived a t by the ordinary methods of
design against flexure combined with direct tension,in the case of straight
walls, and by limiting the overall tensile stress (ignoring reinforcement)
to 200 lb. per square inch. It was essential to provide sufficient reinforcement to take the whole of the tensile stresses. If the silowallswere
designed on that principle, no difficulty would arise in making them
weather-tight.
With regard t o formula (3), it would be noticed in an earlier paragraph
that that formula was intended to refer only to shallow silos and would
not be applicable to a deep silo.
Mr Snows remarks regarding timber construction were quite true and
well recognized.
Steel formwork had been used for continuously moving forms,but they
had the disadvantage that, in general, they weighed and cost more than
wooden forms. They had the advantageof not being affected by changing
moisture and climatic conditions.
in
There was, as Mr Snow had suggested, arisk of distortion taking place
wooden forms, but that was usually taken care of by constructing a rigid
framework a t decking-level between the inner forms of the outer walls.
Provided that that framework was held rigid and kept vertical, the outer
forms couldbe lined up by adjustment of the yoke tie-bolts. Triangulated
tie-bolts were very useful in overcoming any distortion difficulties.
A loud-speakersystem on acontinuouslymoving
job was agreat
advantage, but the Author had notexperienced the necessity of correcting
the levels of the forms by as.much as 1 foot. That would, in most cases,
be a very difficult thing to do, and the forms
should not be allowed to get
so far out of level.
It was agreed that the removal of the formwork from a silo roof was a
difficult and dangerous job. In many cases, the use of steel joists as the
40
DISCUSSION ON THE DESIGN AND

CONSTRUCTION
main members was a great advantage in t8hat direction. Thc suggestion

of the umbrella type of form described was interesting, as was also the
pictorial representation of the progress schedule.
Hexagonal or other queer-shaped binswere used principally to simplify
the distributing and collecting conveyor systems. As pointed out by Mr
Broughton, that was not so important when dealingwith free-flowing
materials, such as grain, where steel spouts could beused to direct the
material on to the conveyors. In the case of silos which were circular in
plan, inter-space binswere useful for storing experimental consignmentsof
the material. Where tapered walls had been used, as in the case of square
silos, theyhad beenfrequentlyreducedfrom,say,
8-10 inches at the
bottom to 44 inches a,t the t,op. For walls of unifornl thickness, 5 inches
was the absoluteminimum that could be used.Anyincrease
on that
made placing of the concrete easier and was worth the small increased cost
of materials.
A spacing of 12-13 feet for yokes, as suggested by Mr Broughton, seemed
to be rather wide. The Author had found that anything more than 8 feet
could cause difficulties.
Mr Broughtons illustrationof a steel yoke (Pig. 25) was interesting, but
it did not matter a great deal what material
was used for the yokes, provided
that they were sufficiently strong. Any design which reduced the weight
of the equipmentgavea
definite advantage. TheAuthorhadnot
exof comperienced any difficulty from theinequalityinthediameter
mercially-rolledjack bars. Cantilevers, supportedfromthetops
of thc
walls, were used to overcome the difficulty of keeping the hanging scaffold
in position after the yokes had been dismantled.
One of the features of continuously moving forms was that the cost of
the workvariedconsiderablyfromjob
to job.There
were so many
factors which influenced the cost, such asthe weather, availabilityof labour,
and site conditions.
The cost of equipment was high, especially at present, and that made it
difficult t o introduce alterations t o equipment which might otherwise bc
anadvantage.
Somescheme of pooling equipment wouldbe
agreat
advantage.
In reply to Mr Manning, the diagram shown in Figs 14 was intended to
give a comparison between results obtained in applying different formulae
to one or twosimpleproblems.
It wouldbe noticed that, in the case
investigated, Airys formula did give higher pressure-values for deep bins
than other formulae, and that result was typical.
Themethod of hoisting forms onchain blocks described was quite
common and was used almost exclusively in America for the construction
of chimneys.
The speeda t which forms couldbe lifted was governed bythe hardening
of the cement. It was essential that theconcrete, when leavingthe bottom
of the forms, should have hardened enough t o stand up without support.
41
The rate of hardening of cementvariedconsiderablyundervarying

atmospheric conditions,and the ratesof progress given bythe Author were
average rates based on average conditions. If higher average rates were
aimed at, a good deal of luck was required in striking ideal conditions, and
much care was needed in choosing and maintaining the consistencyof the
concrete.
The Author agreed entirely with Mr Hey that inter-space bins should
beusedwhereverpossible,
and it was usuallyonly when dealingwith
materials other than grain that thehexagonal or other shape,as an alternative to the circular, was an advantage. The examples
of silos described
andillustrated by M i Hey were extremelyinterestingandformed
a
valuable contribution to the discussion.
The Author agreed with Mr Etches that silos and bunkers generally
were very difficult structures to treat aesthetically ; and that, in order to
overcome the problem, engineerswere induced to introduce ornamentation
which was entirely out of place.
The area occupied by the plan of a set of silos, although important,
seldom had a predominating influence onthe design, because in most cases
site areas were generous.
Mr Etchess curve (Pig. 34,indicating the quantity of steel required in
walls for varying shapes of bins, emphasized the point that the shape did
influence the cost so far as the steel quantities were concerned.
Mr Squiress analysis of the pressures on the bunker slope were quite
correct, but it would be noticed that, even when using his formula, the
normal pressureon the slope, which wasan importantone, was not affected.
When the outlet gate of the silo or bunker was opened and the material
began to flow, the friction between material and the hopper slope broke
down immediately, and consequently the slope of the bottom had to be
greater than the friction angle, otherwise the material would not flow.
The Author had pointed out earlier in the Paper that thevalue of mathematical formulae in the design of bunkers was very limited.
The explanationof varying pressures dueto varying depths of material,
as explained in Pig. 35, was interesting, but in most cases it was advisable
to assume that a bunker was filled t o the top, so that it would not be
overstrained in an emergency.
Mr Fruchtlander had quite rightly pointed out that the Janssen and
Airy formulae could be considered in some ways to be theoretical, but they
were based on practical experiments and the derivation of the formula
which he had given did add considerable value to the discussion.
MADE ANDPRINTEDIN
QREAT BRITAIN BY WILLIAM CLOWES AND SONS, LIMITED,

LONDON AND BEOOLES.

OF SILOS and BUNKERS PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

OF SILOS and BUNKERS PDF

Uploaded by

Copyright:

Available Formats

O F SILOS AKD BUNKERS

Mr Leopold Brook introduced thePaper,inthe

THE DESIGN AND CONSTRUCTION

those forms might distort as much as

Downloaded by [] on [06/07/16]. Copyright ICE Publishing, all rights reserved.

Downloaded by [] on [06/07/16]. Copyright ICE Publishing, all rights reserved.

Downloaded by [] on [06/07/16]. Copyright ICE Publishing, all rights reserved.

Downloaded by [] on [06/07/16]. Copyright ICE Publishing, all rights reserved.

THE DESIGS AND CONSTRUCTION

tended t o become circular,which was therefore the logical form. If

Downloaded by [] on [06/07/16]. Copyright ICE Publishing, all rights reserved.

OF SILOS AXD BUNKERS

Downloaded by [] on [06/07/16]. Copyright ICE Publishing, all rights reserved.

Downloaded by [] on [06/07/16]. Copyright ICE Publishing, all rights reserved.